Instantaneous torque testing method of aero-engine and related system

Abstract

The invention discloses an instantaneous torque testing method of an aero-engine and a related system. The method comprises the following steps: dividing an engine shafting into three temperature zones, and applying acoustic excitation signals with different frequencies to obtain acoustic response signals; determining an operation phase point according to the rotating speed change rate to collect a torque response signal; acquiring inter-axis crosstalk signals to extract a response time difference and an amplitude ratio; collecting pressure field information and extracting spatial distribution and time domain evolution data; an instantaneous torque value is calculated from the signals and data. According to the technical scheme, the problem of torque signal transmission distortion caused by heterogeneous material interface stress wave reflection can be effectively solved, and the accuracy of instantaneous torque measurement is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engine detection and fault diagnosis, and in particular to an aircraft engine instantaneous torque testing method and a related system. Background Art

[0002] The instantaneous torque of an aircraft engine refers to the instantaneous torque transmitted by the shaft system under non-steady-state conditions. This parameter reflects the response characteristics and dynamic stability of the engine to sudden changes in instructions. The measurement of instantaneous torque is crucial to evaluating the performance of an engine: first, it directly reflects the transient characteristics of the engine and can be used to analyze the dynamic response capability of the engine during acceleration and deceleration; second, by monitoring the changing trend of the instantaneous torque, potential fault hazards can be predicted, providing a basis for preventive maintenance of the engine; third, the instantaneous torque data is of great significance for the optimization and verification of the engine control system, and can help improve the reliability and service life of the engine.

[0003] At present, the measurement of instantaneous torque of aircraft engines mainly adopts strain gauge, magnetoelectric and fiber Bragg grating methods. Among them, the strain gauge method indirectly calculates the torque by measuring the surface strain of the shaft system. It has the advantages of simple structure and low cost, but weak anti-interference ability; the magnetoelectric method uses the principle of magnetic flux change and has good dynamic response characteristics, but the stability is insufficient in high temperature environment; the fiber Bragg grating method uses optical fiber sensors to measure the wavelength change caused by torsion. It has the characteristics of strong anti-electromagnetic interference ability, but the system complexity is high. These methods perform well when measuring the torque of homogeneous material structures, but with the increasing use of titanium alloys, composite materials and ceramics and other heterogeneous material integrated structures in modern aircraft engines, a common but long-ignored problem gradually emerges: when the instantaneous torque is transmitted through the interface of heterogeneous materials, stress wave reflection and refraction will occur at the interface, resulting in microsecond delay and distortion of the torque signal in the time domain. The essence of this stress wave reflection phenomenon at the interface of heterogeneous materials is the partial reflection of the stress wave caused by the difference in acoustic impedance of different materials. This phenomenon is particularly significant under high speed and large torque gradient conditions. Since this signal distortion changes dynamically with engine operating conditions and material temperature, it is difficult to eliminate it through simple calibration, thus seriously affecting the accuracy and reliability of instantaneous torque measurement. Summary of the invention

[0004] The main purpose of the present invention is to solve the technical problem that the existing method for measuring the instantaneous torque of an aero-engine has the problem that the stress wave reflection at the interface of heterogeneous materials leads to the distortion of the torque signal transmission.

[0005] A first aspect of the present invention provides a method for testing the instantaneous torque of an aircraft engine. The method for testing the instantaneous torque of an aircraft engine comprises: Dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, wherein the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; Determining an engine operation phase point according to the engine speed change rate, and collecting a phase point torque response signal at the operation phase point; The inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine are collected, and the response time difference and the amplitude ratio between adjacent shafts are extracted according to the inter-axis crosstalk signals to obtain the inter-axis crosstalk characteristic data; Collecting aircraft engine pressure field information, and extracting pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information; The instantaneous torque value of the engine is calculated according to the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

[0006] Preferably, the step of dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, and applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal comprises: The single material area from the air intake end of the engine shaft system to the front of the compressor is divided into the first temperature zone, the double-layer material area from the compressor to the front of the combustion chamber is divided into the second temperature zone, and the multi-layer material area from the rear of the combustion chamber to the turbine is divided into the third temperature zone; Calculating a reference frequency coefficient according to the difference between the idle speed and the maximum operating speed of the engine, multiplying the reference frequency coefficient by the acoustic impedance of the material in the first temperature zone to obtain the sound wave propagation frequency attenuation coefficient in the first temperature zone, setting the frequency of the acoustic excitation signal of the first frequency within the range from the first frequency lower limit value determined by the sound wave propagation frequency attenuation coefficient to 30 kHz, applying the acoustic excitation signal of the first frequency to the first temperature zone and obtaining the first region response amplitude; Calculating a compression frequency coefficient according to a compression ratio of an engine compressor, obtaining a sound wave propagation damping coefficient in the second temperature zone by multiplying the compression frequency coefficient by the acoustic impedance ratio of a material interface in the second temperature zone, setting a frequency of an acoustic excitation signal of a second frequency within a range from a lower limit of the second frequency determined by the acoustic wave propagation damping coefficient to 20 kHz, applying an acoustic excitation signal of a second frequency to the second temperature zone, performing interface transfer compensation on a response signal of the second temperature zone according to a response amplitude in the first zone, and obtaining a response amplitude in the second zone; Calculating a thermoacoustic frequency coefficient according to the temperature of the combustion gas before the engine turbine, obtaining a sound wave propagation velocity coefficient in the third temperature zone by multiplying the thermoacoustic frequency coefficient by the acoustic impedance ratio of the multilayer material in the third temperature zone, setting the frequency of the acoustic excitation signal of the third frequency within the range from the lower limit of the third frequency determined by the sound wave propagation velocity coefficient to 10 kHz, applying an acoustic excitation signal of the third frequency to the third temperature zone, and obtaining a response amplitude in the third zone by combining the response amplitude in the first zone and the response amplitude in the second zone; The first region response amplitude is subjected to frequency compensation to obtain a first acoustic response signal, the second region response amplitude is subjected to material interface compensation to obtain a second acoustic response signal, and the third region response amplitude is subjected to multilayer interface compensation to obtain a third acoustic response signal.

[0007] Preferably, determining the engine operation phase point according to the engine speed change rate, and collecting the phase point torque response signal at the operation phase point, comprises: Obtaining the pressure ratio of each stage of the engine compressor and the expansion ratio of each stage of the turbine, calculating the inter-stage power coupling factor according to the change relationship between the pressure ratios of each stage of the compressor, calculating the inter-stage energy conversion factor according to the change relationship between the expansion ratios of each stage of the turbine, and correlating the inter-stage power coupling factor and the inter-stage energy conversion factor with the engine speed signal to obtain the compensated speed change rate; Collecting the compressor inlet guide vane angle and turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, performing polynomial fitting on the compensation speed change rate, the intake flow correction coefficient and the exhaust flow correction coefficient to obtain the aerodynamic characteristic correction speed change rate; Obtaining the temperature gradient and pressure gradient of each stage of the engine, calculating the temperature gradient ratio and pressure gradient ratio between adjacent stages, performing stress wave transmission compensation operation according to the temperature gradient ratio and pressure gradient ratio, correcting the speed change rate of the aerodynamic characteristic correction, and determining the engine operation phase point; The torque sampling compensation coefficient is calculated according to the change trend of the temperature gradient ratio and the pressure gradient ratio at each level, the shaft system torque signal is collected at the operating phase point, and the sampling signal is corrected according to the torque sampling compensation coefficient to obtain the phase point torque response signal.

[0008] Preferably, collecting the compressor inlet guide vane angle and the turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, and calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, includes: Obtaining the installation angle of each level of stationary blades and the torsion angle of moving blades at the compressor inlet, performing wavelet transform on the installation angle and the torsion angle, extracting the main frequency characteristics and modulation characteristics of the angle change, and calculating the cascade channel area coefficient according to the main frequency characteristics and the modulation characteristics; The static pressure difference and total pressure difference before and after the moving blades of each level at the compressor inlet are collected, and the static pressure difference and total pressure difference are divided into the root area, the middle area and the top area according to the blade height direction. The singular value decomposition method is used to extract the pressure pulsation characteristics of each area, and the blade load distribution coefficient is calculated according to the pressure pulsation characteristics; Adaptively weighting the cascade channel area coefficient and the blade load distribution coefficient, performing instantaneous frequency analysis on the combination result using Hilbert-Huang transform to obtain an inlet flow passage characteristic coefficient, and calculating the inlet Mach number according to the corresponding relationship between the inlet flow passage characteristic coefficient and the inlet guide vane angle; Obtain the throat area of ​​each stage of the turbine and the outlet area of ​​the moving blades, collect the clearance pressure and the circumferential air pressure of each stage of the turbine, form a characteristic matrix with the throat area, outlet area, clearance pressure and circumferential air pressure, and extract the aerodynamic characteristic vector by using the principal component analysis method; Calculating a flow channel convergence factor and an airflow deviation factor according to the aerodynamic characteristic vector, performing modal separation on the flow channel convergence factor and the airflow deviation factor using an empirical mode decomposition method to obtain an exhaust flow channel characteristic coefficient, and calculating a back pressure coefficient according to a corresponding relationship between the exhaust flow channel characteristic coefficient and a turbine nozzle opening; According to the nonlinear mapping relationship between the intake Mach number and the back pressure coefficient, an intake flow correction coefficient and an exhaust flow correction coefficient are obtained respectively.

[0009] Preferably, the inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine are collected, and the response time difference and the amplitude ratio between adjacent shafts are extracted according to the inter-axis crosstalk signals to obtain the inter-axis crosstalk characteristic data, including: Obtain torsional vibration signals and radial displacement signals of the low-pressure shaft, high-pressure shaft and fan shaft, process the torsional vibration signals in sections according to the compressor stages and turbine stages, perform time-frequency decomposition on each section of the signal using continuous wavelet transform, and obtain frequency modulation characteristics of each stage of the shaft system; Acquire radial force signals and axial force signals at adjacent bearings, calculate the force distribution coefficient of the shaft system according to the radial force signals and the axial force signals, perform empirical mode decomposition on the force distribution coefficient of the shaft system, and obtain load distribution characteristics of each level of the shaft system; Performing tensor decomposition on the frequency modulation feature and the load distribution feature, extracting the main mode of shaft system coupling, and calculating the response time difference between adjacent axes according to the main mode of shaft system coupling; Performing Hilbert-Huang transform on the main mode of the shaft system coupling, extracting the shaft system natural frequency characteristics and the load frequency characteristics as instantaneous frequency characteristics, extracting the shaft system amplitude characteristics and the load amplitude characteristics as instantaneous amplitude characteristics, and calculating the amplitude ratio between adjacent shafts according to the instantaneous frequency characteristics and the instantaneous amplitude characteristics; The response time difference and amplitude ratio are jointly analyzed by using variational mode decomposition, and the analysis results are correlated with the clearance variation law of each level of the shaft system to obtain the inter-axis crosstalk characteristic data.

[0010] Preferably, the collecting of the aircraft engine pressure field information and the extraction of the pressure field spatial distribution data and the pressure field time domain evolution data according to the pressure field information include: Obtaining pressure signals at the inlet, interstage and outlet of the engine compressor, performing radial, circumferential and axial orthogonal decomposition on the pressure signals, and calculating the pressure field gradient matrix according to the orthogonal decomposition results; The pressure field gradient matrix is ​​subjected to modal extraction using an inherent orthogonal decomposition method, and the extracted dominant mode is reconstructed according to the energy contribution rate to obtain the spatial distribution data of the main mode; Acquire a first pressure pulsation signal of a stationary blade passage of an engine turbine and a second pressure pulsation signal of a moving blade passage, perform spatiotemporal evolution analysis on the pressure pulsation signals using a dynamic mode decomposition method, and extract a coherent structure of the pressure field; Performing Kriging interpolation operation on the main mode spatial distribution data to obtain pressure field spatial distribution data; A two-dimensional discrete cosine transform is performed on the coherent structure of the pressure field to obtain the time-domain evolution data of the pressure field.

[0011] Preferably, the step of acquiring a first pressure pulsation signal of a stationary blade passage of a turbine engine and a second pressure pulsation signal of a moving blade passage, performing a spatiotemporal evolution analysis on the pressure pulsation signals using a dynamic mode decomposition method, and extracting a coherent structure of the pressure field comprises: Acquire first pressure pulsation signals of the leading edge, trailing edge, pressure surface and suction surface of the stator blades of the high-pressure stage and the low-pressure stage of the turbine, calculate the stator blade passage pressure distribution matrix according to the first pressure pulsation signals, perform singular spectrum decomposition on the stator blade passage pressure distribution matrix, and obtain the stator blade passage pressure characteristic mode; Acquire second pressure pulsation signals of blade crowns, blade basins, and blade backs of turbine high-pressure and low-pressure stages, calculate a blade channel pressure distribution matrix according to the second pressure pulsation signals, perform singular spectrum decomposition on the blade channel pressure distribution matrix, and obtain a blade channel pressure characteristic mode; According to the variation law of the clearance between turbine stages, a nonlinear superposition operation is performed on the stationary blade passage pressure characteristic mode and the moving blade passage pressure characteristic mode to obtain a cascade passage pressure coupling coefficient; The dynamic mode decomposition method is used to perform time series analysis on the pressure coupling coefficient of the cascade channel, and aerodynamic compensation is performed according to the changing relationship between the turbine inlet airflow angle and the outlet airflow angle to obtain the pressure field coherent structure.

[0012] Preferably, the instantaneous torque value of the engine is calculated based on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data, including: The first acoustic response signal, the second acoustic response signal and the third acoustic response signal are divided into a single material segment, a double-layer material segment and a multi-layer material segment, an acoustic wave reflection analysis is performed on the single material segment, an interface transmission analysis is performed on the double-layer material segment, a diffraction characteristic analysis is performed on the multi-layer material segment, and acoustic compensation is performed in combination with the acoustic impedance coefficients of the materials at various levels of the shaft system to obtain an acoustic characteristic weight coefficient; According to the phase point torque response signal and the inter-axle crosstalk characteristic data, the covariance matrix is ​​decomposed, and the energy compensation is performed in combination with the engine compressor efficiency characteristics and turbine efficiency characteristics to obtain the power characteristic weight coefficient; Subspace mapping analysis is performed based on the spatial distribution data of the pressure field and the temporal evolution data of the pressure field, and flow field compensation is performed in combination with the aerodynamic load coefficients of each stage of the turbine and the blade parameters of each stage of the compressor to obtain the aerodynamic characteristic weight coefficient; The acoustic feature weight coefficient, the dynamic feature weight coefficient and the aerodynamic feature weight coefficient are used to construct a tensor set, a deep sparse autoencoder is used to perform feature fusion, and adaptive correction is performed in combination with engine speed and power characteristics to obtain a fusion weight matrix; According to the fusion weight matrix, a weighted fusion operation is performed on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data to obtain the instantaneous torque value of the engine.

[0013] A second aspect of the present invention provides an aircraft engine instantaneous torque test device, the aircraft engine instantaneous torque test device comprising: an acoustic response module, used for dividing the engine shaft system into a first temperature zone, a second temperature zone and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, wherein the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; A phase response module, used to determine an engine operation phase point according to the engine speed change rate, and collect a phase point torque response signal at the operation phase point; A crosstalk analysis module is used to collect inter-axis crosstalk signals of the low-pressure shaft, high-pressure shaft and fan shaft of the aircraft engine, extract the response time difference and amplitude ratio between adjacent shafts according to the inter-axis crosstalk signals, and obtain inter-axis crosstalk characteristic data; A pressure field analysis module is used to collect pressure field information of the aircraft engine and extract pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information; The torque calculation module is used to calculate the instantaneous torque value of the engine according to the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

[0014] The third aspect of the present invention provides an aircraft engine transient torque test system, comprising: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via lines; the at least one processor calls the instructions in the memory so that the aircraft engine transient torque test system executes the steps of the above-mentioned aircraft engine transient torque test method.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the steps of the above-mentioned method for testing the instantaneous torque of an aircraft engine.

[0016] In order to solve the problem of torque signal distortion caused by stress wave reflection at the interface of heterogeneous materials, the present invention proposes a new measurement method. This method solves the problem of stress wave reflection through acoustic partition excitation: since the interface of heterogeneous materials will cause stress wave reflection and refraction, direct measurement of torque signal will produce distortion, so the present invention divides the shaft system into three regions according to the temperature gradient, and applies acoustic excitation of different frequencies to different regions. Through reasonable frequency selection, relatively independent sound wave propagation characteristics can be formed inside each region, so that the influence of the interface of heterogeneous materials on stress wave transmission is confined to a specific area. In this way, the acoustic response signal obtained in each region can accurately reflect the torque transmission characteristics in the region.

[0017] Since the working state of the engine will affect the transmission of the torque signal, the present invention determines the key operating phase points by analyzing the speed change rate and collects the torque response signal at these moments. This method avoids the problem that the traditional fixed sampling method may miss important transient characteristics. The torque signal collected at the phase point can truly reflect the torque change law of the engine under dynamic conditions.

[0018] The multi-axis structure of the engine will cause mutual interference between the axes, affecting the accurate measurement of torque. The present invention simultaneously collects the crosstalk signals of the low-pressure axis, high-pressure axis and fan axis, extracts the time difference and amplitude ratio between adjacent axes, and establishes the characteristic data of crosstalk between axes. These data reveal the interaction law between different axis systems and provide a basis for eliminating the influence of crosstalk between axes.

[0019] In the processing of pressure field data, the present invention obtains the influence of flow field changes on torque transmission by extracting the spatial distribution and time domain evolution characteristics of the pressure field. The pressure field characteristics are directly related to torque transmission and can reflect the stress transfer state at the interface of heterogeneous materials.

[0020] Finally, the present invention fuses the information of the above multiple dimensions to obtain the final instantaneous torque value. This fusion processing mechanism ensures that the real torque signal can be accurately captured under any working conditions, effectively overcoming the signal distortion problem caused by the interface of heterogeneous materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0022] Figure 1A schematic diagram of an embodiment of a method for testing the instantaneous torque of an aircraft engine according to an embodiment of the present invention; Figure 2 A schematic diagram of an embodiment of a transient torque test device for an aircraft engine in an embodiment of the present invention; Figure 3 The figure is a schematic diagram of an embodiment of a transient torque test system for an aircraft engine in an embodiment of the present invention.

[0023] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0026] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, "and / or" in the full text includes three solutions. Taking A and / or B as an example, it includes technical solution A, technical solution B, and technical solution that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, which must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0027] An embodiment of the present application provides a method for testing the instantaneous torque of an aircraft engine. Figure 1 A flow chart of a method for testing the instantaneous torque of an aircraft engine provided in one embodiment of the present application. In this embodiment, the method includes: See also Figure 1, dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; In one embodiment of the present invention, the engine shaft system is divided into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, an acoustic excitation signal of a first frequency is applied to the first temperature zone and a first acoustic response signal is obtained, an acoustic excitation signal of a second frequency is applied to the second temperature zone and a second acoustic response signal is obtained, and an acoustic excitation signal of a third frequency is applied to the third temperature zone and a third acoustic response signal is obtained, comprising: The single material area from the air intake end of the engine shaft system to the front of the compressor is divided into the first temperature zone, the double-layer material area from the compressor to the front of the combustion chamber is divided into the second temperature zone, and the multi-layer material area from the rear of the combustion chamber to the turbine is divided into the third temperature zone; Calculating a reference frequency coefficient according to the difference between the idle speed and the maximum operating speed of the engine, multiplying the reference frequency coefficient by the acoustic impedance of the material in the first temperature zone to obtain the sound wave propagation frequency attenuation coefficient in the first temperature zone, setting the frequency of the acoustic excitation signal of the first frequency within the range from the first frequency lower limit value determined by the sound wave propagation frequency attenuation coefficient to 30 kHz, applying the acoustic excitation signal of the first frequency to the first temperature zone and obtaining the first region response amplitude; Calculating a compression frequency coefficient according to a compression ratio of an engine compressor, obtaining a sound wave propagation damping coefficient in the second temperature zone by multiplying the compression frequency coefficient by the acoustic impedance ratio of a material interface in the second temperature zone, setting a frequency of an acoustic excitation signal of a second frequency within a range from a lower limit of the second frequency determined by the acoustic wave propagation damping coefficient to 20 kHz, applying an acoustic excitation signal of a second frequency to the second temperature zone, performing interface transfer compensation on a response signal of the second temperature zone according to a response amplitude in the first zone, and obtaining a response amplitude in the second zone; Calculating a thermoacoustic frequency coefficient according to the temperature of the combustion gas before the engine turbine, obtaining a sound wave propagation velocity coefficient in the third temperature zone by multiplying the thermoacoustic frequency coefficient by the acoustic impedance ratio of the multilayer material in the third temperature zone, setting the frequency of the acoustic excitation signal of the third frequency within the range from the lower limit of the third frequency determined by the sound wave propagation velocity coefficient to 10 kHz, applying an acoustic excitation signal of the third frequency to the third temperature zone, and obtaining a response amplitude in the third zone by combining the response amplitude in the first zone and the response amplitude in the second zone; The first region response amplitude is subjected to frequency compensation to obtain a first acoustic response signal, the second region response amplitude is subjected to material interface compensation to obtain a second acoustic response signal, and the third region response amplitude is subjected to multilayer interface compensation to obtain a third acoustic response signal.

[0028] The following is a detailed description of the steps involved in the above embodiment: The first step is to divide the engine shaft system into zones. This step uses thermocouple arrays and acoustic detection equipment to scan various parts of the shaft system, and combines the engine structure drawings to determine the precise regional boundaries. The area from the intake end to the front of the compressor is mainly composed of a single titanium alloy or nickel-based high-temperature alloy, with a relatively low temperature (30℃-200℃), and the sound wave propagation characteristics are relatively stable, so it is divided into the first temperature zone; the area from the compressor to the front of the combustion chamber is a double-layer structure composed of an inner layer of high-temperature alloy and an outer layer of thermal insulation material, with a medium temperature (200℃-600℃), and the sound waves will produce a certain degree of reflection and refraction at the material interface, which is divided into the second temperature zone; the area from the combustion chamber to the turbine is composed of multiple layers of materials such as high-temperature alloy substrate, thermal barrier coating, thermal insulation layer and cooling structure, with the highest temperature (600℃-1500℃), and the complexity of sound wave propagation is the highest, which is divided into the third temperature zone. This division method based on material composition and temperature gradient can specifically solve the problem of different sound wave propagation characteristics caused by material composition and temperature differences in different regions, making subsequent acoustic analysis more accurate.

[0029] The second step is to perform acoustic excitation and response acquisition on the first temperature zone. This step first reads the idle speed (e.g., 4000rpm) and the maximum operating speed (e.g., 12000rpm) from the engine control unit (ECU), calculates the difference to be 8000rpm, and multiplies the difference by the acoustic conversion factor (0.002Hz / rpm) to obtain the reference frequency coefficient of 16Hz. Then, an acoustic impedance analyzer is used to measure the acoustic impedance value (27.3× kg / m²s), multiplied by the reference frequency coefficient, the sound wave propagation frequency attenuation coefficient is 0.437× . According to this coefficient, the lower limit of the first frequency is determined to be 26.2kHz, and 28kHz is selected as the excitation frequency within the range from this lower limit to 30kHz. An acoustic excitation signal is applied to the first temperature zone using a piezoelectric transducer array with precise frequency control, and the response signal is collected by a high-sensitivity piezoelectric sensor arranged on the surface of the shaft system. The response amplitude of the first area is obtained through a high-speed data acquisition system and spectrum analysis. The high-frequency range (26.2kHz-30kHz) is selected because the temperature in the first temperature zone is low and the material is single, and high-frequency sound waves can provide higher measurement resolution and improve torque measurement accuracy.

[0030] The third step is to perform acoustic excitation and response acquisition on the second temperature zone. This step first obtains the compressor pressure ratio (e.g. 8:1) from the engine monitoring system and substitutes it into the acoustic compression coefficient calculation formula =0.15×(PR-1), where Represents the compression frequency coefficient, PR represents the compressor pressure ratio, and 0.15 is an empirical constant used to convert the pressure ratio change into the acoustic frequency influencing factor. The compression frequency coefficient is 1.05. Then the interface acoustic analyzer is used to measure the acoustic impedance ratio (1.62) of the inner and outer materials of the second temperature zone (such as Inconel 718 and thermal insulation ceramics), and multiplied by the compression frequency coefficient to obtain the acoustic wave propagation damping coefficient of 1.701. According to this coefficient, the lower limit of the second frequency is determined to be 15.3kHz, and 17kHz is selected as the excitation frequency within the range of this lower limit to 20kHz. An intermediate frequency piezoelectric transducer is used to apply an acoustic excitation signal to the second temperature zone, and the response signal is collected through a distributed acoustic sensing network. The digital signal processor is combined with the first region response amplitude obtained in the first step, and the interface transfer function algorithm is applied to eliminate the signal distortion caused by the interface reflection to obtain the second region response amplitude. The intermediate frequency range (15.3kHz-20kHz) is selected because there is a material interface in the second temperature zone, and the intermediate frequency sound wave can maintain sufficient penetration and provide sufficient resolution, balancing the measurement depth and accuracy requirements.

[0031] The fourth step is to perform acoustic excitation and response acquisition on the third temperature zone. In this step, the temperature sensor array is used to obtain the gas temperature before the turbine (for example, 1100°C) and substitute it into the thermoacoustic conversion equation: =0.004×(T-800), where Represents the thermoacoustic frequency coefficient, T represents the gas temperature before the turbine (℃), 0.004 is the thermoacoustic conversion coefficient, and 800 is the reference temperature (℃), which is used to calculate the influence of high temperature on the propagation characteristics of sound waves. The thermoacoustic frequency coefficient is calculated to be 1.2. Then, a multi-layer acoustic scanning system is used to measure the comprehensive acoustic impedance ratio (2.83) of each layer of material in the third temperature zone (such as high-temperature alloy, ceramic thermal barrier coating, and thermal insulation layer), and multiply it with the thermoacoustic frequency coefficient to obtain the sound wave propagation velocity coefficient of 3.396. According to this coefficient, the lower limit of the third frequency is determined to be 6.8kHz, and 8kHz is selected as the excitation frequency within the range from this lower limit to 10kHz. An acoustic excitation signal is applied to the third temperature zone using a high-temperature resistant low-frequency piezoelectric transducer, and the response signal is collected through a high-temperature resistant acoustic sensor network. Combined with the response amplitudes obtained in the first two steps, a multi-region signal correlation processor is used for comprehensive analysis to obtain the response amplitude of the third region. The low-frequency range (6.8kHz-10kHz) is selected because the third temperature zone has high temperature and complex materials. Low-frequency sound waves have strong penetrating ability and can pass through the interface of multiple layers of materials to obtain complete acoustic information, ensuring that reliable measurement results can still be obtained in high-temperature complex material areas.

[0032] The fifth step is to compensate the response amplitude of each region. In this step, a dedicated acoustic signal processing system is used to apply the frequency response function to the response amplitude of the first region for frequency compensation to eliminate the amplitude distortion caused by the uneven frequency response in the high-frequency band; the acoustic impedance matching algorithm is applied to the response amplitude of the second region for material interface compensation to correct the signal attenuation and phase shift caused by the reflection and refraction of the double-layer material interface; the multi-layer acoustic propagation model is applied to the response amplitude of the third region for multi-layer interface compensation to correct the response distortion caused by the complex interference of multi-layer materials. In the compensation process, an iterative optimization algorithm is used to ensure the optimal selection of compensation parameters, and finally the first acoustic response signal, the second acoustic response signal and the third acoustic response signal reflecting the true torque transmission characteristics of each temperature zone are obtained. This differentiated compensation processing method fully considers the differences in material properties and acoustic propagation characteristics in different temperature zones, avoids the problem of measurement error accumulation caused by the traditional single processing method, and improves the overall measurement accuracy.

[0033] Please continue reading Figure 1 , determining an engine operation phase point according to the engine speed change rate, and collecting a phase point torque response signal at the operation phase point; In one embodiment of the present invention, determining the engine operation phase point according to the engine speed change rate, and collecting the phase point torque response signal at the operation phase point, comprises: Obtaining the pressure ratio of each stage of the engine compressor and the expansion ratio of each stage of the turbine, calculating the inter-stage power coupling factor according to the change relationship between the pressure ratios of each stage of the compressor, calculating the inter-stage energy conversion factor according to the change relationship between the expansion ratios of each stage of the turbine, and correlating the inter-stage power coupling factor and the inter-stage energy conversion factor with the engine speed signal to obtain the compensated speed change rate; Collecting the compressor inlet guide vane angle and turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, performing polynomial fitting on the compensation speed change rate, the intake flow correction coefficient and the exhaust flow correction coefficient to obtain the aerodynamic characteristic correction speed change rate; Obtaining the temperature gradient and pressure gradient of each stage of the engine, calculating the temperature gradient ratio and pressure gradient ratio between adjacent stages, performing stress wave transmission compensation operation according to the temperature gradient ratio and pressure gradient ratio, correcting the speed change rate of the aerodynamic characteristic correction, and determining the engine operation phase point; The torque sampling compensation coefficient is calculated according to the change trend of the temperature gradient ratio and the pressure gradient ratio at each level, the shaft system torque signal is collected at the operating phase point, and the sampling signal is corrected according to the torque sampling compensation coefficient to obtain the phase point torque response signal.

[0034] The following is a detailed description of the steps involved in the above embodiment: The first step is to obtain the parameters of each stage of the engine and calculate the coupling factor. This step collects the pressure ratio data of each stage (for example, 1.8:1 for the first stage, 1.6:1 for the second stage, and 1.4:1 for the third stage) through the pressure sensor array installed between the compressor stages, and collects the expansion ratio data of each stage (for example, 2.5:1 for the high pressure stage and 2.0:1 for the low pressure stage) through the temperature and pressure sensors of the turbine stages. When calculating the inter-stage power coupling factor, first calculate the pressure ratio difference between adjacent compressor stages, divide it by the pressure ratio of the previous stage, and then sum all stages and multiply it by the power conversion constant (value 0.8). When calculating the inter-stage energy conversion factor, divide the product of the expansion ratios of adjacent turbine stages by their sum, sum all stages and multiply them by the energy conversion constant (value 1.2). Finally, the rate of change of the engine speed is combined with these two factors, specifically the speed change rate multiplied by (1 plus the power coupling factor minus the energy conversion factor) to obtain the compensated speed change rate. This method of considering the relationship between the compressor and turbine stages can accurately reflect the power transfer characteristics in the multi-stage compression and expansion process and improve the accuracy of the speed change rate.

[0035] The second step is to collect aerodynamic parameters and make corrections. This step collects the compressor inlet guide vane angle (for example, 23°) and the turbine nozzle opening (for example, 85%) through the position sensor. When calculating the intake flow correction coefficient, the correction value is obtained using a linear relationship based on the deviation between the guide vane angle and the optimal guide vane angle (usually 20°). Specifically, it is the product of the reference value 0.95 minus the deviation value and the coefficient 0.005. When calculating the exhaust flow correction coefficient, the correction value is obtained using a linear relationship based on the difference between the nozzle opening and the nominal nozzle opening (usually 80%). Specifically, it is the reference value 0.9 plus the product of the difference and the coefficient 0.002. Then, a polynomial fitting program is used to weight the compensation speed change rate, the intake flow correction coefficient, and the exhaust flow correction coefficient. The weight coefficient is determined by experiment to obtain the aerodynamic characteristic correction speed change rate. This multi-parameter fitting method integrates the influence of intake and exhaust characteristics on speed changes and more comprehensively reflects the dynamic aerodynamic characteristics of the engine.

[0036] The third step is to obtain gradient information and determine the operating phase point. This step obtains the temperature gradient (such as 120℃ / stage) and pressure gradient (such as 200kPa / stage) data of each stage of the engine through the temperature sensor and pressure sensor array. Calculate the temperature gradient ratio between adjacent stages (that is, the temperature gradient of the current stage divided by the temperature gradient of the next stage) and the pressure gradient ratio (that is, the pressure gradient of the current stage divided by the pressure gradient of the next stage). When performing stress wave transmission compensation, the aerodynamic characteristic correction speed change rate is multiplied by the compensation factor, which is obtained by summing the temperature gradient ratio and the pressure gradient ratio multiplied by their respective weight coefficients. Use numerical analysis software to find the inflection point or extreme point of the final correction speed change rate to determine the engine operation phase point. This method that considers temperature and pressure gradients can capture the changing characteristics of stress waves during the transmission process of different material interfaces and more accurately determine the optimal time for torque sampling.

[0037] The fourth step is to calculate the compensation coefficient and collect the torque signal. When calculating the torque sampling compensation coefficient in this step, the change rate of the temperature gradient ratio and the pressure gradient ratio (that is, the degree of their change over time) is considered, and the change rate of the temperature gradient ratio of each stage is multiplied by the corresponding influence coefficient and then summed, and then the sum of the change rate of the pressure gradient ratio of each stage multiplied by the corresponding influence coefficient is added, and finally the reference value 1 is added to obtain the compensation coefficient. At the determined operating phase point, a high-precision torque sensor is used to collect the original torque signal of the shaft system, and the signal is multiplied by the calculated torque sampling compensation coefficient to obtain the phase point torque response signal. This compensation method based on the gradient change trend can effectively eliminate the torque signal distortion caused by the reflection of stress waves at the interface of heterogeneous materials, and improve the accuracy of instantaneous torque measurement, especially under non-steady-state conditions such as rapid acceleration and deceleration of the engine.

[0038] In one embodiment of the present invention, the collecting of the compressor inlet guide vane angle and the turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, and calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, comprises: Obtaining the installation angle of each level of stationary blades and the torsion angle of moving blades at the compressor inlet, performing wavelet transform on the installation angle and the torsion angle, extracting the main frequency characteristics and modulation characteristics of the angle change, and calculating the cascade channel area coefficient according to the main frequency characteristics and the modulation characteristics; The static pressure difference and total pressure difference before and after the moving blades of each level at the compressor inlet are collected, and the static pressure difference and total pressure difference are divided into the root area, the middle area and the top area according to the blade height direction. The singular value decomposition method is used to extract the pressure pulsation characteristics of each area, and the blade load distribution coefficient is calculated according to the pressure pulsation characteristics; Adaptively weighting the cascade channel area coefficient and the blade load distribution coefficient, performing instantaneous frequency analysis on the combination result using Hilbert-Huang transform to obtain an inlet flow passage characteristic coefficient, and calculating the inlet Mach number according to the corresponding relationship between the inlet flow passage characteristic coefficient and the inlet guide vane angle; Obtain the throat area of ​​each stage of the turbine and the outlet area of ​​the moving blades, collect the clearance pressure and the circumferential air pressure of each stage of the turbine, form a characteristic matrix with the throat area, outlet area, clearance pressure and circumferential air pressure, and extract the aerodynamic characteristic vector by using the principal component analysis method; Calculating a flow channel convergence factor and an airflow deviation factor according to the aerodynamic characteristic vector, performing modal separation on the flow channel convergence factor and the airflow deviation factor using an empirical mode decomposition method to obtain an exhaust flow channel characteristic coefficient, and calculating a back pressure coefficient according to a corresponding relationship between the exhaust flow channel characteristic coefficient and a turbine nozzle opening; According to the nonlinear mapping relationship between the intake Mach number and the back pressure coefficient, an intake flow correction coefficient and an exhaust flow correction coefficient are obtained respectively.

[0039] The following is a detailed description of the steps involved in the above embodiment: The first step is to obtain the installation angle of the compressor stator blades and the torsion angle of the moving blades and calculate the cascade channel area coefficient. In this step, the installation angle of the compressor inlet stator blades (for example, 40° for the first stage and 38° for the second stage) and the torsion angle of the moving blades (for example, 12° for the first stage and 15° for the second stage) are obtained by an optical angle measuring instrument. The digital signal processor is used to perform discrete wavelet transform on these angle data, and the Daubechies wavelet is used as the basis function, and the decomposition level is set to 5. By analyzing the wavelet coefficients, the main frequency characteristics of the angle change (representing the frequency component of periodic changes, such as 7.5Hz related to the speed, 230Hz of the blade passing frequency) and the modulation characteristics (representing the modulation frequency of the amplitude change, such as 2.3Hz related to the intake pulsation) are extracted. The main frequency characteristics reflect the change in the flow channel area caused by the periodic swing of the blades, and the modulation characteristics reflect the contraction or expansion of the flow channel caused by non-periodic disturbances. According to these characteristics, the weighted calculation method is used to obtain the cascade channel area coefficient, which represents the ratio of the effective area of ​​the flow channel between the blades to the design area. This wavelet analysis-based method can accurately capture the dynamic change characteristics of blade angles and reflect the real-time flow capacity of the cascade channel.

[0040] The second step is to collect and analyze the pressure difference information before and after the compressor moving blades to calculate the load distribution coefficient. This step collects static pressure difference (such as 4.2 kPa for the first stage and 3.8 kPa for the second stage) and total pressure difference (such as 15.6 kPa for the first stage and 14.2 kPa for the second stage) data through the micro pressure sensor array installed before and after each stage of the compressor moving blades. The collected pressure difference data is divided into three regions according to the blade height direction: the root region accounting for 0-30% of the blade height, the middle region accounting for 30-70% of the blade height, and the top region accounting for 70-100% of the blade height. A time series matrix is ​​constructed for the pressure data of each region, and the matrix is ​​decomposed using the singular value decomposition method. The first few singular values ​​and their corresponding singular vectors that account for more than 90% of the total energy are retained to extract the pressure pulsation characteristics of each region. The size of the singular value reflects the intensity of the pressure pulsation, and the singular vector reflects the spatial distribution pattern of the pulsation. Based on these characteristics, the blade load distribution coefficient is calculated, which represents the relative distribution of the load in each region of the blade. This multi-region analysis method can accurately characterize the radial distribution unevenness of blade loads and effectively identify complex aerodynamic phenomena such as local flow separation and secondary flow.

[0041] The third step is to combine and analyze and calculate the intake Mach number. In this step, the cascade channel area coefficient and the blade load distribution coefficient are input into the adaptive weighting algorithm, and the weights of the two coefficients are automatically adjusted according to the current engine operating conditions to obtain the combined result. The adaptive weight is calculated in real time by the engine control system based on parameters such as speed and power to ensure that the two coefficients are given a reasonable degree of importance under different operating conditions. The Hilbert-Huang transform is applied to the combined result to decompose the signal into multiple intrinsic mode functions and extract the instantaneous frequency characteristics. The Hilbert-Huang transform is a method suitable for nonlinear and non-stationary signal analysis, which can effectively handle the rapid changes in the airflow parameters of aircraft engines. The intake flow channel characteristic coefficient is obtained based on the instantaneous frequency characteristics, and the intake Mach number is calculated based on the calibration relationship curve between the coefficient and the inlet guide vane angle (obtained in advance through wind tunnel tests). This adaptive combination and advanced signal processing method significantly improves the accuracy of the intake Mach number calculation and meets the full operating conditions of the engine.

[0042] The fourth step is to obtain the turbine aerodynamic parameters and construct a characteristic matrix. This step obtains the throat area of ​​the turbine stator blades at each stage (such as 400mm² for high-pressure stage and 600mm² for low-pressure stage) and the outlet area of ​​the moving blades (such as 450mm² for high-pressure stage and 700mm² for low-pressure stage) through engine design data and real-time measurement. At the same time, the clearance pressure of each stage of the turbine (such as 800kPa for high-pressure stage and 400kPa for low-pressure stage) and the air pressure data of multiple points around the wheel (one measuring point every 45° along the circumference) are collected through the pressure sensor array. These data are organized into a characteristic matrix, where the rows of the matrix represent different measurement points and the columns represent different parameters. The principal component analysis method is applied to the characteristic matrix to calculate the covariance matrix and solve its eigenvalues ​​and eigenvectors. The first few principal components with a contribution rate of more than 95% are selected as aerodynamic eigenvectors. Principal component analysis can effectively reduce the data dimension, extract key information, and eliminate redundancy and noise interference between parameters. This multi-parameter comprehensive analysis method can fully capture the geometric characteristics and aerodynamic performance of the turbine flow channel.

[0043] The fifth step is to analyze the aerodynamic eigenvector and calculate the back pressure coefficient. This step calculates the flow channel convergence factor (indicating the rate of change of the flow channel cross-sectional area along the flow direction) and the airflow deviation factor (indicating the degree of deviation between the actual airflow direction and the design direction) based on the aerodynamic eigenvector. The empirical mode decomposition method is applied to these two factors to decompose the signal into multiple intrinsic mode functions and a residual trend term, and extract the modes directly related to the turbine aerodynamic performance. Empirical mode decomposition is an adaptive signal processing method suitable for processing nonlinear and non-stationary aerodynamic signals. The exhaust flow channel characteristic coefficient is obtained by reconstructing the key modes, which comprehensively reflects the flow capacity of the exhaust system. The back pressure coefficient is calculated based on the calibration relationship curve between the coefficient and the turbine nozzle opening (obtained in advance through the high-temperature aerodynamic test bench). This multi-step analysis method can accurately characterize the complex aerodynamic characteristics of the exhaust system and adapt to different engine exhaust configurations.

[0044] The sixth step is to calculate the flow correction coefficient based on the intake and exhaust parameters. This step uses the numerical simulation results to establish a nonlinear mapping relationship between the intake Mach number and the back pressure coefficient. Through table lookup or interpolation methods, the intake flow correction coefficient and the exhaust flow correction coefficient are obtained respectively according to the currently calculated intake Mach number and back pressure coefficient. The nonlinear mapping relationship takes into account many factors such as gas compressibility, flow channel geometry, and gas state equation, and can accurately reflect the complex relationship between intake and exhaust flow and Mach number and back pressure. This correction method based on the physical model can significantly improve the accuracy of flow estimation and provide reliable aerodynamic basic data for instantaneous torque measurement.

[0045] Please continue reading Figure 1 , collecting the inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine, extracting the response time difference and the amplitude ratio between adjacent shafts according to the inter-axis crosstalk signals, and obtaining the inter-axis crosstalk characteristic data; In one embodiment of the present invention, the inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine are collected, and the response time difference and the amplitude ratio between adjacent shafts are extracted according to the inter-axis crosstalk signals to obtain the inter-axis crosstalk characteristic data, including: Obtain torsional vibration signals and radial displacement signals of the low-pressure shaft, high-pressure shaft and fan shaft, process the torsional vibration signals in sections according to the compressor stages and turbine stages, perform time-frequency decomposition on each section of the signal using continuous wavelet transform, and obtain frequency modulation characteristics of each stage of the shaft system; Acquire radial force signals and axial force signals at adjacent bearings, calculate the force distribution coefficient of the shaft system according to the radial force signals and the axial force signals, perform empirical mode decomposition on the force distribution coefficient of the shaft system, and obtain load distribution characteristics of each level of the shaft system; Performing tensor decomposition on the frequency modulation feature and the load distribution feature, extracting the main mode of shaft system coupling, and calculating the response time difference between adjacent axes according to the main mode of shaft system coupling; Performing Hilbert-Huang transform on the main mode of the shaft system coupling, extracting the shaft system natural frequency characteristics and the load frequency characteristics as instantaneous frequency characteristics, extracting the shaft system amplitude characteristics and the load amplitude characteristics as instantaneous amplitude characteristics, and calculating the amplitude ratio between adjacent shafts according to the instantaneous frequency characteristics and the instantaneous amplitude characteristics; The response time difference and amplitude ratio are jointly analyzed by using variational mode decomposition, and the analysis results are correlated with the clearance variation law of each level of the shaft system to obtain the inter-axis crosstalk characteristic data.

[0046] The following is a detailed description of the steps involved in the above embodiment: The first step is to obtain and process the shaft vibration signal. This step collects torsional vibration signals and radial displacement signals by installing optical torsional vibration sensors and eddy current displacement sensors on the low-pressure shaft, high-pressure shaft and fan shaft of the aircraft engine. The torsional vibration signal reflects the angular displacement change of the shaft system during rotation, and the radial displacement signal reflects the radial swing of the shaft. The collected signal is processed in segments according to the number of stages of the engine compressor and turbine. For example, if a certain type of engine contains 9 compressors and 4 turbines, the signal is divided into 13 segments. The continuous wavelet transform is used to perform time-frequency decomposition on each segment of the signal. The specific operation is to select the Mexican hat wavelet as the mother wavelet, set the scale parameter to 1-64, and the translation step size to 1. The wavelet coefficients are calculated for each segment of the signal to generate a time-frequency energy distribution diagram. The frequency components with the strongest signal energy and their changes over time are extracted from the time-frequency diagram to obtain the frequency modulation characteristics. For example, a certain segment of the high-pressure shaft signal may be found to have a base frequency of 120Hz and obvious speed-related amplitude modulation. This analysis method based on continuous wavelet transform can accurately capture the frequency variation characteristics of the shaft system during the transmission process between different stages, reflecting the influence of the interface of heterogeneous materials on torque transmission.

[0047] The second step is to obtain and process the bearing force signal. In this step, the radial force signal (x-direction and y-direction) and the axial force signal (z-direction) are collected by the strain sensor array installed at each bearing seat of the engine. These force signals reflect the support reaction force received by the shaft system during operation. According to the collected force signals, the magnitude and direction of the resultant force at each bearing are calculated, and the force distribution coefficient of the shaft system is calculated considering the relative position between the bearings. This coefficient represents the uniformity of force distribution on the shaft system. The calculation method is to divide the force value at each bearing by the average value of all bearing force values ​​to obtain a series of dimensionless coefficients. The empirical mode decomposition method is applied to these shaft system force distribution coefficients to decompose the signal into multiple intrinsic mode functions and a residual trend. Empirical mode decomposition is an adaptive signal processing method that can decompose complex signals into components with clear physical meanings. The main modes related to the dynamic characteristics of the shaft system are identified from the intrinsic mode functions to obtain the load distribution characteristics of each level of the shaft system. For example, for the high-pressure shaft of a certain type of engine, it may be found that the first mode represents the basic load distribution, and the third mode reflects the periodic fluctuation of the bearing load. This multi-dimensional force signal analysis method can comprehensively capture the dynamic characteristics of the shaft system and provide accurate load distribution information for inter-axis crosstalk analysis.

[0048] The third step is to perform tensor decomposition to extract the main mode. This step organizes the frequency modulation features and load distribution features obtained in the above two steps into a three-dimensional tensor. The three dimensions of the tensor represent the type of axis, the position on the axis, and the feature type. The tensor is decomposed by Tucker to decompose it into the product of the core tensor and three factor matrices. Several tensor elements with the largest contribution rate are selected from the decomposition results to reconstruct the main mode of shaft system coupling. These main modes represent the intrinsic relationship between frequency characteristics and load distribution. Based on the time response curve of the main mode, the response time difference between different axes is calculated. For example, the main mode time series of the high-pressure axis and the low-pressure axis are cross-correlated to find the peak position of the cross-correlation function. The time difference corresponding to this position is the response time difference between the two axes. This multimodal analysis method based on tensor decomposition effectively integrates vibration characteristics and load characteristics, and can capture the coupling relationship between different shaft systems.

[0049] The fourth step is to perform signal transformation and calculate the amplitude ratio. In this step, the Hilbert-Huang transform is performed on the main mode of the shaft system coupling. First, the signal is empirically decomposed to obtain the intrinsic mode function, and then the Hilbert transform is performed on each mode function to calculate the instantaneous frequency and instantaneous amplitude. From the transformation results, the shaft system intrinsic frequency characteristics (such as 80Hz for the low-pressure shaft and 150Hz for the high-pressure shaft) and the load frequency characteristics (such as 45Hz for the bearing and 320Hz for the blade) are extracted as the instantaneous frequency characteristics, and the shaft system amplitude characteristics and load amplitude characteristics are extracted as the instantaneous amplitude characteristics. According to the instantaneous amplitude ratio of the corresponding frequencies of the adjacent shafts, the amplitude ratio is calculated. For example, at a frequency of 120Hz, the amplitude ratio of the high-pressure shaft to the low-pressure shaft is 1.8:1, indicating that the transmission intensity of the corresponding torque signal of the high-pressure shaft at this frequency is 1.8 times that of the low-pressure shaft. This time-frequency analysis method based on the Hilbert-Huang transform can accurately characterize the energy transfer relationship between different shaft systems and provide a quantitative basis for the crosstalk analysis between shafts.

[0050] The fifth step is to analyze the characteristics of crosstalk between axes. This step uses variational mode decomposition to jointly analyze the response time difference and amplitude ratio. Variational mode decomposition is a non-recursive adaptive modal analysis method that can effectively handle nonlinear and non-stationary signals. The number of modes is set to 3-5, the bandwidth constraint is 50Hz, and a series of modes with limited bandwidth are decomposed through an iterative optimization algorithm. The analysis results are correlated with the change rules of the clearances at various levels of the shaft system obtained from the engine speed sensor and the clearance sensor to obtain the crosstalk characteristic data between axes. This crosstalk characteristic data comprehensively reflects the dynamic coupling relationship between different shaft systems, and can effectively eliminate the influence of mutual interference in multi-axis systems on instantaneous torque measurement.

[0051] Please continue reading Figure 1 , collecting aircraft engine pressure field information, and extracting pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information; In one embodiment of the present invention, the collecting of aircraft engine pressure field information and the extraction of pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information include: Obtaining pressure signals at the inlet, interstage and outlet of the engine compressor, performing radial, circumferential and axial orthogonal decomposition on the pressure signals, and calculating the pressure field gradient matrix according to the orthogonal decomposition results; The pressure field gradient matrix is ​​subjected to modal extraction using an inherent orthogonal decomposition method, and the extracted dominant mode is reconstructed according to the energy contribution rate to obtain the spatial distribution data of the main mode; Acquire a first pressure pulsation signal of a stationary blade passage of an engine turbine and a second pressure pulsation signal of a moving blade passage, perform spatiotemporal evolution analysis on the pressure pulsation signals using a dynamic mode decomposition method, and extract a coherent structure of the pressure field; Performing Kriging interpolation operation on the main mode spatial distribution data to obtain pressure field spatial distribution data; A two-dimensional discrete cosine transform is performed on the coherent structure of the pressure field to obtain the time-domain evolution data of the pressure field.

[0052] The following is a detailed description of the steps involved in the above embodiment: The first step is to obtain and decompose the pressure signal. This step collects pressure signals through an array of high-frequency pressure sensors installed at the inlet, gaps between stages, and outlet of the aircraft engine compressor. The specific arrangement is: 8 sensors are evenly distributed circumferentially at the inlet and 3 layers (inner ring, middle ring, outer ring) radially, and the same arrangement is made between stages. There are 12 sensors circumferentially and 4 layers radially at the outlet, totaling about 100 measuring points. The acquisition frequency is set to 20kHz to ensure that high-frequency pressure fluctuations are captured. The acquired pressure signal data is subjected to three-dimensional orthogonal decomposition: first decomposed into a combination of basis functions and coefficients along the radial direction (from the inner wall to the outer wall), then Fourier decomposition is performed along the circumferential direction (0°-360°) to obtain harmonic components of each order, and finally polynomial decomposition is performed along the axial direction (from the inlet to the outlet). The pressure gradients in each direction are calculated based on the decomposition results and organized into a pressure field gradient matrix. Each element of the matrix represents the pressure gradient vector at a specific position and time. For example, during the acceleration of a certain type of engine, the radial pressure gradient recorded at the second-stage compressor position was 2.5 kPa / cm, the circumferential gradient was 0.8 kPa / radian, and the axial gradient was 5.2 kPa / cm. This three-dimensional orthogonal decomposition method can fully characterize the spatial distribution characteristics of the pressure field inside the engine and capture local flow characteristics and large-scale structures.

[0053] The second step is to extract the dominant modes of the pressure field. This step uses the inherent orthogonal decomposition method (also known as principal component analysis or Karhunen-Loève decomposition) to process the pressure field gradient matrix. First, the autocorrelation matrix is ​​constructed, and the eigenvalues ​​and eigenvectors of the matrix are calculated. The size of the eigenvalue represents the energy contribution of each mode, and the eigenvector represents the spatial structure of the corresponding mode. According to the eigenvalue, the first few modes (usually 5-8) with a cumulative energy contribution rate of 95% are selected as the dominant modes. For example, the analysis results of a certain type of engine show that the energy contribution rate of the first mode is 48%, which is manifested as the average distribution of the overall pressure field; the contribution rate of the second mode is 27%, which is manifested as the axial pressure gradient change; the contribution rate of the third mode is 12%, which is manifested as the first-order non-uniform distribution in the circumferential direction. These dominant modes are weighted and reconstructed according to their respective energy contribution rates to obtain the spatial distribution data of the main modes. This mode extraction method based on energy distribution can effectively reduce the data dimension, retain the most physically meaningful large-scale structure in the pressure field, and filter out small-scale noise and disturbances.

[0054] The third step is to analyze the turbine pressure pulsation signal. In this step, the pressure pulsation signal is collected by high-temperature pressure sensors installed in the turbine stator and moving blade channels. The sensors of the stator channel are arranged at 12 positions, including the leading edge, trailing edge, pressure surface and suction surface, and the sensors of the moving blade channel are arranged at 9 positions, including the blade crown, blade basin and blade back. The collected first pressure pulsation signal (stator) and second pressure pulsation signal (moving blade) form a spatiotemporal data matrix, in which the rows of the matrix represent different spatial positions and the columns represent different time points. The dynamic mode decomposition method is applied to these matrices: first, the time translation pair of the data matrix is ​​constructed, then the linear mapping operator between the two is calculated, and finally the operator is eigendecomposed to obtain the eigenvalue (representing the time evolution characteristics) and the eigenvector (representing the spatial distribution characteristics). The key dynamic modes are identified from the decomposition results, and the coherent structure of the pressure field expressed as a coherent spatiotemporal structure is extracted. For example, in the analysis of a certain type of engine, it was found that there was a periodic pressure fluctuation structure dominated by the blade passing frequency at the outlet of the turbine stator, which caused a resonant response at the inlet of the moving blade. This method based on dynamic mode decomposition can effectively identify the main dynamic structures in the pressure field and reveal the propagation law of pressure waves in the blade channel.

[0055] The fourth step is to perform Kriging interpolation to calculate the spatial distribution. This step performs Kriging interpolation on the spatial distribution data of the main mode, and expands the data of discrete measurement points to the entire flow field space. First, a spatial variogram model is established, and the Gaussian variogram is selected as the basic model. For each spatial position point, the optimal linear unbiased estimate is obtained by solving the Kriging equations according to the data and position of the known measurement points. The entire flow field is gridded (such as 100 points in the axial direction, 50 points in the radial direction, and 72 points in the circumferential direction), and each grid point is interpolated to obtain high-resolution pressure field spatial distribution data. For example, the analysis results of a certain type of engine show that there is a high-pressure area in the high-pressure compressor outlet area, and the pressure value is 10% higher than that of the surrounding area. The area presents a 30° circumferential fan-shaped distribution. This geostatistical Kriging interpolation method has the characteristics of optimal estimation, and can generate a continuous and smooth pressure field distribution map based on limited measurement point data, providing detailed flow field background information for torque analysis.

[0056] The fifth step is to perform discrete cosine transform to obtain time domain evolution data. This step performs a two-dimensional discrete cosine transform on the coherent structure of the pressure field to convert the spatial domain information into a frequency domain representation. First, the coherent structure is organized into a two-dimensional matrix according to the spatial position, and then the two-dimensional discrete cosine transform algorithm is applied to obtain the transformation coefficient matrix. The first 20% of the coefficients with the largest absolute values ​​after the transformation are selected, and the remaining coefficients are set to zero, and then the inverse transform is performed to obtain the compressed pressure field time domain evolution data. For example, the analysis results of a certain type of engine show that: during the acceleration process, the pressure wave in the turbine propagates from the stationary blades to the moving blades at a speed of 17m / s, and the amplitude is attenuated by about 25% during the propagation process. This data compression method based on discrete cosine transform can not only retain the main characteristics of the time domain evolution of the pressure field, but also effectively reduce the amount of data, which is convenient for subsequent instantaneous torque calculation.

[0057] In one embodiment of the present invention, the step of acquiring a first pressure pulsation signal of a stationary blade passage of an engine turbine and a second pressure pulsation signal of a moving blade passage, performing a spatiotemporal evolution analysis on the pressure pulsation signals using a dynamic mode decomposition method, and extracting a coherent structure of a pressure field includes: Acquire first pressure pulsation signals of the leading edge, trailing edge, pressure surface and suction surface of the stator blades of the high-pressure stage and the low-pressure stage of the turbine, calculate the stator blade passage pressure distribution matrix according to the first pressure pulsation signals, perform singular spectrum decomposition on the stator blade passage pressure distribution matrix, and obtain the stator blade passage pressure characteristic mode; Acquire second pressure pulsation signals of blade crowns, blade basins, and blade backs of turbine high-pressure and low-pressure stages, calculate a blade channel pressure distribution matrix according to the second pressure pulsation signals, perform singular spectrum decomposition on the blade channel pressure distribution matrix, and obtain a blade channel pressure characteristic mode; According to the variation law of the clearance between turbine stages, a nonlinear superposition operation is performed on the stationary blade passage pressure characteristic mode and the moving blade passage pressure characteristic mode to obtain a cascade passage pressure coupling coefficient; The dynamic mode decomposition method is used to perform time series analysis on the pressure coupling coefficient of the cascade channel, and aerodynamic compensation is performed according to the changing relationship between the turbine inlet airflow angle and the outlet airflow angle to obtain the pressure field coherent structure.

[0058] The following is a detailed description of the steps involved in the above embodiment: The first step is to obtain and process the pressure pulsation signal of the stator blade. This step collects the first pressure pulsation signal through the high-temperature pressure sensors installed at the key positions of the stator blades of the high-pressure stage and the low-pressure stage of the turbine. The specific arrangement is: install a pressure sensor on the leading edge (the front end of the windward surface), the trailing edge (the trailing edge where the airflow flows out), the pressure surface (convex surface, the side with higher airflow pressure) and the suction surface (concave surface, the side with lower airflow pressure) of each stator blade, and collect 8 points at the high-pressure stage and the low-pressure stage, for a total of 16 measurement points. The acquisition frequency is set to 40kHz, and the sampling time is 10 seconds to ensure that the complete pressure pulsation cycle is captured. According to the collected first pressure pulsation signal, the stator channel pressure distribution matrix is ​​constructed, and the rows of the matrix represent different spatial positions and the columns represent different time points. The matrix is ​​subjected to singular spectrum decomposition to calculate the singular values ​​of the matrix and the corresponding left and right singular vectors. The size of the singular value represents the energy contribution of each mode, the left singular vector represents the spatial mode, and the right singular vector represents the time evolution characteristics. The modes corresponding to the first 3-5 singular values ​​are selected as the characteristic modes of the stator passage pressure. For example, in the high-pressure turbine of a certain type of engine, the first characteristic mode is the average distribution of the overall pressure, accounting for 65% of the energy; the second characteristic mode is the pressure difference distribution between the pressure surface and the suction surface, accounting for 22% of the energy; the third characteristic mode is the pressure fluctuation between the leading edge and the trailing edge, accounting for 8% of the energy. This method based on singular spectrum decomposition can effectively extract the most physically meaningful pressure distribution mode in the stator passage, reflecting the characteristics of blade load and airflow distribution.

[0059] The second step is to obtain and process the pressure pulsation signal of the moving blade. In this step, the high-temperature pressure sensors installed at the key positions of the moving blades of the high-pressure stage and the low-pressure stage of the turbine collect the second pressure pulsation signal. The specific arrangement is: a pressure sensor is installed on each moving blade crown (the outermost part of the blade opposite to the casing), blade basin (concave surface, the main surface of airflow impact) and blade back (convex surface, the surface of airflow outflow), and 6 points are collected at the high-pressure stage and the low-pressure stage, totaling 12 measurement points. The acquisition frequency is also set to 40kHz, and it is collected synchronously with the stationary blade signal. According to the collected second pressure pulsation signal, the moving blade channel pressure distribution matrix is ​​constructed, and the matrix structure is similar to that of the stationary blade. The matrix is ​​also subjected to singular spectrum decomposition to extract the characteristic mode of the moving blade channel pressure. For example, in the analysis of the moving blades of the same engine, the first characteristic mode is manifested as the pressure distribution in the blade height direction, accounting for 58% of the energy; the second characteristic mode is manifested as the pressure difference distribution between the blade basin and the blade back, accounting for 25% of the energy; the third characteristic mode is manifested as the pressure pulsation near the blade crown, accounting for 10% of the energy. This pressure analysis of key parts of the moving blades reveals the distribution of unsteady aerodynamic loads on the moving blades during high-speed rotation, which helps to understand the dynamic characteristics during torque transmission.

[0060] The third step is to calculate the blade channel pressure coupling coefficient. This step first obtains the change law of the turbine interstage gap through the speed sensor and the gap measurement system. During the operation of the engine, the gap will change with factors such as speed and temperature. For example, during the starting stage, the gap gradually decreases from 0.8mm in the cold state to 0.5mm in the hot state. According to this change law, the static blade channel pressure characteristic mode and the moving blade channel pressure characteristic mode are nonlinearly superimposed. The specific operation is: the two characteristic modes are connected according to the spatial position mapping relationship. Considering the transfer characteristics of the airflow between the static blade and the moving blade, a time delay factor is introduced, which corresponds to the time required for the airflow to flow from the static blade to the moving blade (usually 0.2-0.5 milliseconds). The polynomial weighted method is used for nonlinear superposition, and the weighting coefficient is dynamically adjusted with the change of the interstage gap. The smaller the gap, the greater the coupling strength. Through this processing, the blade channel pressure coupling coefficient is obtained, which characterizes the aerodynamic coupling strength between the static blade and the moving blade. This nonlinear superposition method that considers the dynamic change of the interstage gap can accurately reflect the complex aerodynamic interference effect between the static blade and the moving blade, and capture the key characteristics in the torque transmission process.

[0061] The fourth step is to perform dynamic modal decomposition analysis to obtain the pressure field coherent structure. This step uses the dynamic modal decomposition method to perform time series analysis on the pressure coupling coefficient of the cascade channel. First, a time evolution matrix is ​​constructed, and the coefficients are arranged in chronological order, with each column representing the state at a certain moment. A mapping relationship between the states at the previous and next moments is established, the optimal linear mapping matrix is ​​calculated, and the matrix is ​​eigendecomposed to obtain eigenvalues ​​(representing the time evolution characteristics) and eigenvectors (representing the spatial distribution characteristics). At the same time, the change relationship between the turbine inlet airflow angle (the angle with the axial direction, usually 20°-40°) and the outlet airflow angle (usually 60°-80°) is obtained through the flow field measurement system. According to the change of the airflow angle, the aerodynamic correction coefficient is calculated to compensate for the amplitude and phase of the dynamic mode. Finally, the aerodynamically compensated pressure field coherent structure is obtained, which characterizes the large-scale dynamic characteristics with physical correlation in the pressure field. For example, in the analysis of a certain type of engine, it was found that during the acceleration process, there is a pressure wave structure that propagates along the flow direction with the blade passing frequency as the main frequency. This structure starts from the high-pressure stator blades, passes through the moving blades, the low-pressure stator blades, and finally to the low-pressure moving blades, showing obvious cascade propagation characteristics. This dynamic analysis method can reveal the temporal and spatial evolution of the pressure field and provide an important basis for understanding the torque transmission mechanism in complex flow fields.

[0062] Please continue reading Figure 1 The instantaneous torque value of the engine is calculated based on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

[0063] In one embodiment of the present invention, the instantaneous torque value of the engine is calculated based on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data, including: The first acoustic response signal, the second acoustic response signal and the third acoustic response signal are divided into a single material segment, a double-layer material segment and a multi-layer material segment, an acoustic wave reflection analysis is performed on the single material segment, an interface transmission analysis is performed on the double-layer material segment, a diffraction characteristic analysis is performed on the multi-layer material segment, and acoustic compensation is performed in combination with the acoustic impedance coefficients of the materials at various levels of the shaft system to obtain an acoustic characteristic weight coefficient; According to the phase point torque response signal and the inter-axle crosstalk characteristic data, the covariance matrix is ​​decomposed, and the energy compensation is performed in combination with the engine compressor efficiency characteristics and turbine efficiency characteristics to obtain the power characteristic weight coefficient; Subspace mapping analysis is performed based on the spatial distribution data of the pressure field and the temporal evolution data of the pressure field, and flow field compensation is performed in combination with the aerodynamic load coefficients of each stage of the turbine and the blade parameters of each stage of the compressor to obtain the aerodynamic characteristic weight coefficient; The acoustic feature weight coefficient, the dynamic feature weight coefficient and the aerodynamic feature weight coefficient are used to construct a tensor set, a deep sparse autoencoder is used to perform feature fusion, and adaptive correction is performed in combination with engine speed and power characteristics to obtain a fusion weight matrix; According to the fusion weight matrix, a weighted fusion operation is performed on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data to obtain the instantaneous torque value of the engine.

[0064] The following is a detailed description of the steps involved in the above embodiment: The first step is to analyze and compensate the acoustic response signal. This step first uses the shafting structure diagram and material distribution information to divide the first acoustic response signal, the second acoustic response signal, and the third acoustic response signal into a single material segment (such as a titanium alloy segment), a double-layer material segment (such as a high-temperature alloy and a thermal insulation layer), and a multi-layer material segment (such as a multi-layer composite material segment) according to the material composition. The pulse echo method is used to perform acoustic wave reflection analysis on the single material segment. By emitting pulses and receiving reflected echoes, the propagation speed and attenuation characteristics of the sound wave in the material are calculated, and the acoustic wave reflection coefficient is extracted (for example, the reflection coefficient of the titanium alloy segment is 0.15). The transmission wave technology is used to perform interface transmission analysis on the double-layer material segment. By measuring the amplitude ratio and phase difference of the transmission wave, the transmission coefficient and reflection coefficient of the interface are calculated (for example, the transmission coefficient of the interface between the high-temperature alloy and the ceramic thermal insulation layer is 0.65). The scattered wave technology is used to analyze the diffraction characteristics of the multi-layer material segment, measure the scattered wave intensity distribution in different directions, and calculate the acoustic wave diffraction index of the material (for example, the diffraction index of the multi-layer material in the turbine area is 0.38). Combined with the acoustic impedance coefficients of various shafting materials obtained from the material database (such as titanium alloy 27.3× kg / m²s, nickel-based superalloy 41.5× kg / m²s), and the acoustic transfer matrix method is used to perform acoustic compensation for the entire system to obtain the acoustic characteristic weight coefficient, which reflects the intensity distribution of the acoustic response in different material segments. This segmented analysis method based on material composition differentiates the acoustic characteristics of different material structures and effectively solves the problem of different acoustic wave propagation characteristics at the interface of heterogeneous materials.

[0065] The second step is to process the torque response and inter-axis crosstalk data to obtain the power characteristic weight. This step organizes the phase point torque response signal and the inter-axis crosstalk characteristic data into a data matrix, where the rows of the matrix represent different shaft system positions and the columns represent different time points. The covariance matrix is ​​calculated for the matrix, and then the eigenvalue decomposition is performed to extract the main eigenvectors as the power characteristic basis. The compressor efficiency characteristics (such as the isentropic efficiency of each compressor at the design point is 84%-89%) and the turbine efficiency characteristics (such as the isentropic efficiency of each turbine at the design point is 88%-92%) are obtained from the engine performance database. According to these efficiency characteristics, the energy conversion efficiency of each stage is calculated, and energy compensation is performed considering the deviation between the actual and ideal working conditions. The specific operation is to correct the amplitude of the power characteristic basis to make it more consistent with the actual energy transfer characteristics. Finally, the power characteristic weight coefficient is obtained, which characterizes the transmission characteristics of torque in different shaft system sections. For example, the analysis results of a certain type of engine show that the power weight of the low-pressure shaft section is 0.32, the power weight of the high-pressure shaft section is 0.45, and the power weight of the fan shaft section is 0.23. This compensation method based on actual efficiency characteristics can accurately reflect the energy conversion efficiency of each engine component and correct the torque estimation error caused by ignoring efficiency changes in traditional measurements.

[0066] The third step is to process the pressure field data to obtain the aerodynamic feature weight. This step uses the pressure field spatial distribution data and the pressure field time domain evolution data to construct the pressure field feature space. The kernel principal component analysis method is used for subspace mapping analysis to map the high-dimensional pressure field data to the low-dimensional feature space and extract the main aerodynamic features. The aerodynamic load coefficients of each stage of the turbine (such as the load coefficient of the first-stage turbine blade is 1.2) and the blade parameters of each stage of the compressor (such as the angle of attack of the first-stage compressor blade is 4° and the torsion angle is 12°) are obtained from the engine aerodynamic design data. According to these parameters, combined with the computational fluid dynamics model, the aerodynamic correction coefficients under actual working conditions are calculated, and the subspace mapping results are compensated for the flow field. Finally, the aerodynamic feature weight coefficient is obtained, which characterizes the intensity of the influence of aerodynamic characteristics on torque transmission. For example, during the acceleration process of a certain engine, the aerodynamic weight of the compressor section is 0.28, the aerodynamic weight of the combustion chamber section is 0.15, and the aerodynamic weight of the turbine section is 0.57. This compensation method that takes actual aerodynamic load into consideration can accurately characterize the contribution of aerodynamic force to shaft system torque, thus improving the comprehensiveness and accuracy of torque measurement.

[0067] The fourth step is to construct a tensor set and perform feature fusion. In this step, the acoustic feature weight coefficient, the dynamic feature weight coefficient, and the aerodynamic feature weight coefficient are organized into a three-dimensional tensor set. The three dimensions of the tensor represent the feature type, the axis position, and the time point, respectively. A deep sparse autoencoder based on the TensorFlow framework is used for feature fusion. The autoencoder structure is a 5-layer network (input layer-encoding layer 1-encoding layer 2-decoding layer-output layer), and the number of neurons in each layer is 100-64-32-64-100. The L1 regularization constraint is introduced in the encoding layer to achieve sparse representation of features. The Adam optimizer is used for training, the learning rate is set to 0.001, and the number of iterations is 5000. By minimizing the reconstruction error and the sparsity penalty term, nonlinear dimensionality reduction and feature extraction of the tensor set are achieved. Combined with the speed (such as N1=12000rpm, N2=23000rpm) and power characteristics (such as the output power is 15MW) obtained from the engine control system, the fusion result is adaptively corrected to obtain the fusion weight matrix. This deep learning-based feature fusion method can adaptively extract the optimal feature combination from multi-source heterogeneous data, effectively solving the problem that traditional fusion methods are difficult to handle high-dimensional nonlinear data.

[0068] The fifth step is to calculate the instantaneous torque by weighted fusion. This step uses the fusion weight matrix to perform weighted fusion operations on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axis crosstalk feature data, the pressure field spatial distribution data, and the pressure field time domain evolution data. The specific implementation is to weight each signal according to its corresponding weight in the fusion weight matrix, consider the time delay between different signals (such as the propagation delay between the acoustic signal and the pressure signal is about 0.5-2 milliseconds), and align the signal phase. The instantaneous torque value of the engine is obtained by weighted superposition operation. For example, during the rapid acceleration of a certain type of engine, the instantaneous torque calculated by this method quickly rises from the initial 2000N·m to 6500N·m, with a response time of 0.8 seconds. Compared with the traditional measurement method, the dynamic response error is reduced by 65%, and the torque peak measurement accuracy is improved by 42%. This multi-dimensional information fusion method makes full use of information from multiple angles such as acoustics, dynamics and aerodynamics, effectively overcomes the measurement error caused by stress wave reflection at the interface of heterogeneous materials, and improves the accuracy and reliability of instantaneous torque measurement of aircraft engines.

[0069] The instantaneous torque test method of the aircraft engine in the embodiment of the present invention is described above. The instantaneous torque test device of the aircraft engine in the embodiment of the present invention is described below. Figure 2 In one embodiment of the present invention, an instantaneous torque test device for an aircraft engine includes: an acoustic response module 101, for dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, wherein the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; A phase response module 102, for determining an engine operation phase point according to an engine speed change rate, and collecting a phase point torque response signal at the operation phase point; The crosstalk analysis module 103 is used to collect the inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine, extract the response time difference and the amplitude ratio between adjacent shafts according to the inter-axis crosstalk signals, and obtain the inter-axis crosstalk characteristic data; The pressure field analysis module 104 is used to collect the pressure field information of the aircraft engine, and extract the pressure field spatial distribution data and the pressure field time domain evolution data according to the pressure field information; The torque calculation module 105 is used to calculate the instantaneous torque value of the engine according to the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

[0070] above Figure 2 The instantaneous torque testing device of the aircraft engine in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The instantaneous torque testing system of the aircraft engine in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0071] Figure 3 2 is a schematic diagram of the structure of an instantaneous torque test system for an aircraft engine provided by an embodiment of the present invention. The instantaneous torque test system 200 for an aircraft engine may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 210 (for example, one or more processors) and a memory 220, and one or more storage media 230 (for example, one or more mass storage device terminals) storing application programs 233 or data 232. Among them, the memory 220 and the storage medium 230 may be temporary storage or permanent storage. The program stored in the storage medium 230 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the instantaneous torque test system 200 for an aircraft engine. Furthermore, the processor 210 may be configured to communicate with the storage medium 230, and execute a series of instruction operations in the storage medium 230 on the instantaneous torque test system 200 for an aircraft engine to implement the steps of the instantaneous torque test method for an aircraft engine.

[0072] The aircraft engine instantaneous torque test system 200 may further include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input and output interfaces 260, and / or one or more operating systems 231, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 3 The structure of the aircraft engine transient torque test system shown does not constitute a limitation on the aircraft engine transient torque test system provided by the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0073] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the instantaneous torque test method of the aircraft engine.

[0074] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0075] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0076] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A method for testing the instantaneous torque of an aircraft engine, characterized in that: include: Dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, wherein the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; Determining an engine operation phase point according to the engine speed change rate, and collecting a phase point torque response signal at the operation phase point; The inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine are collected, and the response time difference and the amplitude ratio between adjacent shafts are extracted according to the inter-axis crosstalk signals to obtain the inter-axis crosstalk characteristic data; Collecting aircraft engine pressure field information, and extracting pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information; The instantaneous torque value of the engine is calculated according to the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

2. The method for testing the instantaneous torque of an aircraft engine according to claim 1, characterized in that: The method of dividing the engine shaft system into a first temperature zone, a second temperature zone, and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, and applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal comprises: The single material area from the air intake end of the engine shaft system to the front of the compressor is divided into the first temperature zone, the double-layer material area from the compressor to the front of the combustion chamber is divided into the second temperature zone, and the multi-layer material area from the rear of the combustion chamber to the turbine is divided into the third temperature zone; Calculating a reference frequency coefficient according to the difference between the idle speed and the maximum operating speed of the engine, multiplying the reference frequency coefficient by the acoustic impedance of the material in the first temperature zone to obtain the sound wave propagation frequency attenuation coefficient in the first temperature zone, setting the frequency of the acoustic excitation signal of the first frequency within the range from the first frequency lower limit value determined by the sound wave propagation frequency attenuation coefficient to 30 kHz, applying the acoustic excitation signal of the first frequency to the first temperature zone and obtaining the first region response amplitude; Calculating a compression frequency coefficient according to a compression ratio of an engine compressor, obtaining a sound wave propagation damping coefficient in the second temperature zone by multiplying the compression frequency coefficient by the acoustic impedance ratio of a material interface in the second temperature zone, setting a frequency of an acoustic excitation signal of a second frequency within a range from a lower limit of the second frequency determined by the acoustic wave propagation damping coefficient to 20 kHz, applying an acoustic excitation signal of a second frequency to the second temperature zone, performing interface transfer compensation on a response signal of the second temperature zone according to a response amplitude in the first zone, and obtaining a response amplitude in the second zone; Calculating a thermoacoustic frequency coefficient according to the temperature of the combustion gas before the engine turbine, obtaining a sound wave propagation velocity coefficient in the third temperature zone by multiplying the thermoacoustic frequency coefficient by the acoustic impedance ratio of the multilayer material in the third temperature zone, setting the frequency of the acoustic excitation signal of the third frequency within the range from the lower limit of the third frequency determined by the sound wave propagation velocity coefficient to 10 kHz, applying an acoustic excitation signal of the third frequency to the third temperature zone, and obtaining a response amplitude in the third zone by combining the response amplitude in the first zone and the response amplitude in the second zone; The first region response amplitude is subjected to frequency compensation to obtain a first acoustic response signal, the second region response amplitude is subjected to material interface compensation to obtain a second acoustic response signal, and the third region response amplitude is subjected to multilayer interface compensation to obtain a third acoustic response signal.

3. The instantaneous torque test method of an aircraft engine according to claim 1, characterized in that: Determining the engine operation phase point according to the engine speed change rate, and collecting the phase point torque response signal at the operation phase point, includes: Obtaining the pressure ratio of each stage of the engine compressor and the expansion ratio of each stage of the turbine, calculating the inter-stage power coupling factor according to the change relationship between the pressure ratios of each stage of the compressor, calculating the inter-stage energy conversion factor according to the change relationship between the expansion ratios of each stage of the turbine, and correlating the inter-stage power coupling factor and the inter-stage energy conversion factor with the engine speed signal to obtain the compensated speed change rate; Collecting the compressor inlet guide vane angle and turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, performing polynomial fitting on the compensation speed change rate, the intake flow correction coefficient and the exhaust flow correction coefficient to obtain the aerodynamic characteristic correction speed change rate; Obtaining the temperature gradient and pressure gradient of each stage of the engine, calculating the temperature gradient ratio and pressure gradient ratio between adjacent stages, performing stress wave transmission compensation operation according to the temperature gradient ratio and pressure gradient ratio, correcting the speed change rate of the aerodynamic characteristic correction, and determining the engine operation phase point; The torque sampling compensation coefficient is calculated according to the change trend of the temperature gradient ratio and the pressure gradient ratio at each level, the shaft system torque signal is collected at the operating phase point, and the sampling signal is corrected according to the torque sampling compensation coefficient to obtain the phase point torque response signal.

4. The method for testing the instantaneous torque of an aircraft engine according to claim 3, characterized in that: The collecting of the compressor inlet guide vane angle and the turbine nozzle opening, calculating the intake flow correction coefficient according to the corresponding relationship between the guide vane angle and the intake Mach number, and calculating the exhaust flow correction coefficient according to the corresponding relationship between the nozzle opening and the back pressure coefficient, includes: Obtaining the installation angle of each level of stationary blades and the torsion angle of moving blades at the compressor inlet, performing wavelet transform on the installation angle and the torsion angle, extracting the main frequency characteristics and modulation characteristics of the angle change, and calculating the cascade channel area coefficient according to the main frequency characteristics and the modulation characteristics; The static pressure difference and total pressure difference before and after the moving blades of each level at the compressor inlet are collected, and the static pressure difference and total pressure difference are divided into the root area, the middle area and the top area according to the blade height direction. The singular value decomposition method is used to extract the pressure pulsation characteristics of each area, and the blade load distribution coefficient is calculated according to the pressure pulsation characteristics; Adaptively weighting the cascade channel area coefficient and the blade load distribution coefficient, performing instantaneous frequency analysis on the combination result using Hilbert-Huang transform to obtain an inlet flow passage characteristic coefficient, and calculating the inlet Mach number according to the corresponding relationship between the inlet flow passage characteristic coefficient and the inlet guide vane angle; Obtain the throat area of ​​each stage of the turbine and the outlet area of ​​the moving blades, collect the clearance pressure and the circumferential air pressure of each stage of the turbine, form a characteristic matrix with the throat area, outlet area, clearance pressure and circumferential air pressure, and extract the aerodynamic characteristic vector by using the principal component analysis method; Calculating a flow channel convergence factor and an airflow deviation factor according to the aerodynamic characteristic vector, performing modal separation on the flow channel convergence factor and the airflow deviation factor using an empirical mode decomposition method to obtain an exhaust flow channel characteristic coefficient, and calculating a back pressure coefficient according to a corresponding relationship between the exhaust flow channel characteristic coefficient and a turbine nozzle opening; According to the nonlinear mapping relationship between the intake Mach number and the back pressure coefficient, an intake flow correction coefficient and an exhaust flow correction coefficient are obtained respectively.

5. The method for testing the instantaneous torque of an aircraft engine according to claim 1, characterized in that: The inter-axis crosstalk signals of the low-pressure shaft, the high-pressure shaft and the fan shaft of the aircraft engine are collected, and the response time difference and the amplitude ratio between adjacent shafts are extracted according to the inter-axis crosstalk signals to obtain the inter-axis crosstalk characteristic data, including: Obtain torsional vibration signals and radial displacement signals of the low-pressure shaft, high-pressure shaft and fan shaft, process the torsional vibration signals in sections according to the compressor stages and turbine stages, perform time-frequency decomposition on each section of the signal using continuous wavelet transform, and obtain frequency modulation characteristics of each stage of the shaft system; Acquire radial force signals and axial force signals at adjacent bearings, calculate the force distribution coefficient of the shaft system according to the radial force signals and the axial force signals, perform empirical mode decomposition on the force distribution coefficient of the shaft system, and obtain load distribution characteristics of each level of the shaft system; Performing tensor decomposition on the frequency modulation feature and the load distribution feature, extracting the main mode of shaft system coupling, and calculating the response time difference between adjacent axes according to the main mode of shaft system coupling; Performing Hilbert-Huang transform on the main mode of the shaft system coupling, extracting the shaft system natural frequency characteristics and the load frequency characteristics as instantaneous frequency characteristics, extracting the shaft system amplitude characteristics and the load amplitude characteristics as instantaneous amplitude characteristics, and calculating the amplitude ratio between adjacent shafts according to the instantaneous frequency characteristics and the instantaneous amplitude characteristics; The response time difference and amplitude ratio are jointly analyzed by using variational mode decomposition, and the analysis results are correlated with the clearance variation law of each level of the shaft system to obtain the inter-axis crosstalk characteristic data.

6. The method for testing the instantaneous torque of an aircraft engine according to claim 1, characterized in that: The collecting of the aircraft engine pressure field information and the extraction of the pressure field spatial distribution data and the pressure field time domain evolution data according to the pressure field information include: Obtaining pressure signals at the inlet, interstage and outlet of the engine compressor, performing radial, circumferential and axial orthogonal decomposition on the pressure signals, and calculating the pressure field gradient matrix according to the orthogonal decomposition results; The pressure field gradient matrix is ​​subjected to modal extraction using an inherent orthogonal decomposition method, and the extracted dominant mode is reconstructed according to the energy contribution rate to obtain the spatial distribution data of the main mode; Acquire a first pressure pulsation signal of a stationary blade passage of an engine turbine and a second pressure pulsation signal of a moving blade passage, perform spatiotemporal evolution analysis on the pressure pulsation signals using a dynamic mode decomposition method, and extract a coherent structure of the pressure field; Performing Kriging interpolation operation on the main mode spatial distribution data to obtain pressure field spatial distribution data; A two-dimensional discrete cosine transform is performed on the coherent structure of the pressure field to obtain the time-domain evolution data of the pressure field.

7. The method for testing the instantaneous torque of an aircraft engine according to claim 6, characterized in that: The first pressure pulsation signal of the stationary blade passage of the engine turbine and the second pressure pulsation signal of the moving blade passage are obtained, and the temporal and spatial evolution analysis of the pressure pulsation signal is performed by using a dynamic mode decomposition method to extract the coherent structure of the pressure field, including: Acquire first pressure pulsation signals of the leading edge, trailing edge, pressure surface and suction surface of the stator blades of the high-pressure stage and the low-pressure stage of the turbine, calculate the stator blade passage pressure distribution matrix according to the first pressure pulsation signals, perform singular spectrum decomposition on the stator blade passage pressure distribution matrix, and obtain the stator blade passage pressure characteristic mode; Acquire second pressure pulsation signals of blade crowns, blade basins, and blade backs of turbine high-pressure and low-pressure stages, calculate a blade channel pressure distribution matrix according to the second pressure pulsation signals, perform singular spectrum decomposition on the blade channel pressure distribution matrix, and obtain a blade channel pressure characteristic mode; According to the variation law of the clearance between turbine stages, a nonlinear superposition operation is performed on the stationary blade passage pressure characteristic mode and the moving blade passage pressure characteristic mode to obtain a cascade passage pressure coupling coefficient; The dynamic mode decomposition method is used to perform time series analysis on the pressure coupling coefficient of the cascade channel, and aerodynamic compensation is performed according to the changing relationship between the turbine inlet airflow angle and the outlet airflow angle to obtain the pressure field coherent structure.

8. The method for testing the instantaneous torque of an aircraft engine according to claim 1, characterized in that: The instantaneous torque value of the engine is calculated based on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data, including: The first acoustic response signal, the second acoustic response signal and the third acoustic response signal are divided into a single material segment, a double-layer material segment and a multi-layer material segment, an acoustic wave reflection analysis is performed on the single material segment, an interface transmission analysis is performed on the double-layer material segment, a diffraction characteristic analysis is performed on the multi-layer material segment, and acoustic compensation is performed in combination with the acoustic impedance coefficients of the materials at various levels of the shaft system to obtain an acoustic characteristic weight coefficient; According to the phase point torque response signal and the inter-axle crosstalk characteristic data, the covariance matrix is ​​decomposed, and the energy compensation is performed in combination with the engine compressor efficiency characteristics and turbine efficiency characteristics to obtain the power characteristic weight coefficient; Subspace mapping analysis is performed based on the spatial distribution data of the pressure field and the temporal evolution data of the pressure field, and flow field compensation is performed in combination with the aerodynamic load coefficients of each stage of the turbine and the blade parameters of each stage of the compressor to obtain the aerodynamic characteristic weight coefficient; The acoustic feature weight coefficient, the dynamic feature weight coefficient and the aerodynamic feature weight coefficient are used to construct a tensor set, a deep sparse autoencoder is used to perform feature fusion, and adaptive correction is performed in combination with engine speed and power characteristics to obtain a fusion weight matrix; According to the fusion weight matrix, a weighted fusion operation is performed on the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data to obtain the instantaneous torque value of the engine.

9. An instantaneous torque test device for an aircraft engine, characterized in that: include: an acoustic response module, used for dividing the engine shaft system into a first temperature zone, a second temperature zone and a third temperature zone with temperatures from low to high, applying an acoustic excitation signal of a first frequency to the first temperature zone and obtaining a first acoustic response signal, applying an acoustic excitation signal of a second frequency to the second temperature zone and obtaining a second acoustic response signal, applying an acoustic excitation signal of a third frequency to the third temperature zone and obtaining a third acoustic response signal, wherein the first frequency is greater than the second frequency, and the second frequency is greater than the third frequency; A phase response module, used to determine an engine operation phase point according to the engine speed change rate, and collect a phase point torque response signal at the operation phase point; A crosstalk analysis module is used to collect inter-axis crosstalk signals of the low-pressure shaft, high-pressure shaft and fan shaft of the aircraft engine, extract the response time difference and amplitude ratio between adjacent shafts according to the inter-axis crosstalk signals, and obtain inter-axis crosstalk characteristic data; A pressure field analysis module is used to collect pressure field information of the aircraft engine and extract pressure field spatial distribution data and pressure field time domain evolution data according to the pressure field information; The torque calculation module is used to calculate the instantaneous torque value of the engine according to the first acoustic response signal, the second acoustic response signal, the third acoustic response signal, the phase point torque response signal, the inter-axle crosstalk characteristic data, the pressure field spatial distribution data and the pressure field time domain evolution data.

10. An aircraft engine instantaneous torque test system, characterized in that: The instantaneous torque test system of the aircraft engine comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the aircraft engine transient torque testing system to execute the steps of the aircraft engine transient torque testing method as described in any one of claims 1-8.

Images